Deep Learning in Healthcare Summit schedule

Your.MD: Revolutionising pre-primary care access via chatbot

Matteo Berlucchi will delve into the process of building an artificial medical brain that is capable of finding the most likely cause of an indisposition. Through a set of intelligent algorithms, based on a doctors’ thought process, Your.MD is capable of reaching a comparable result to a doctor when given the same information -  totally personalised to fit an individual's history, profile and current situation. Matteo will address the fundamental components that make this possible: a medical knowledge base, a chat-based interface, a medical AI brain and a safe platform to store individualised structural medical data.

Matteo Berlucchi is CEO at Your.MD - the world’s first Personal Health Assistant to combine Artificial Intelligence, machine learning and content from the NHS to deliver free, personalised health information to a global audience. As one of the first Internet entrepreneurs in Europe, Berlucchi has spent his career building and growing companies and services operating across the online and mobile sectors. Prior to Your.MD, Berlucchi has served as Chief Digital Officer at Northern & Shell, founder and CEO of the ‘social ebook’ platform aNobii, as well as the first live news platform on iPhone, Livestation, and Skinkers, an award-winning digital communication platform. Berlucchi has also served as an advisor to multiple individuals and businesses.

Hiba is a Director at Doctorpreneurs, the global community for medical entrepreneurs designed to inspire, connect and accelerate healthcare professionals with an interest in medical innovation and technology. Doctorpreneurs aim to provide medical entrepreneurs with the insight, network and opportunities to lead innovative improvement in healthcare.

Hiba is a graduate of Imperial College Business School and is studying medicine at Imperial College London. Her current research with her team under Lord Darzi focuses on medication noncompliance and development of mAdherence solutions. She is Co-Founder of the MedTech UK platform which is currently growing as a national university network. Hiba is working on projects in healthcare policy and digital health, such as developing an engagement tool for frontline clinical staff with Innovation 4 Health and has previously supported 3M Health Care in building an application in clinical education.

Codec helps brands remain relevant to their target audiences. Our content intelligence platform combines big social data with versatile machine learning solutions to tell brands what content resonates with the audiences they want to target, all before the brand spends big on high-production content. The result is interesting content, engaged consumers and increased marketing ROI. We enable brands to start contributing to culture through more relevant, valuable and impactful content. In this talk I will describe our innovative approach to one of the foundations in our research - brand audience analysis and relevant tribe search.

Ekaterina Volkova-Volkmar is a lead data scientist at Codec. Prior to this role she was a data scientist in the field of digital health. She finished her PhD at the Max Planck Institute for Biological Cybernetics in Tübingen, Germany, in 2014. With research background in neuroscience, computer science, and computational linguistics, Ekaterina is interested in integrating artificial intelligence into digital solutions to scale human expertise and excellence.

Alejandro (Alex) Jaimes is CTO & Chief Scientist at Acesio. Acesio focuses on Big Data for predictive analytics in Healthcare to tackle disease at worldwide scale, impacting individuals and entire populations. We use Artificial Intelligence to collect and analyze vast quantities of data to track and predict disease in ways that have never been done before- leveraging environmental variables, population movements, sensor data, and the web. Prior to joining Acesio, Alex was CTO at AiCure and prior to that he was Director of Research/Video Product at Yahoo where he led research and contributions to Yahoo's video products, managing teams of scientists and engineers in New York City, Sunnyvale, Bangalore, and Barcelona. His work focuses on Machine Learning, mixing qualitative and quantitative methods to gain insights on user behavior for product innovation. He has published widely in the top-tier conferences (KDD, WWW, RecSys, CVPR, ACM Multimedia, etc), has been a visiting professor (KAIST), and is a frequent speaker at international academic and industry events. He is a scientist and innovator with 15+ years of international experience in research leading to product impact (Yahoo, KAIST, Telefonica, IDIAP-EPFL, Fuji Xerox, IBM, Siemens, and AT&T Bell Labs). He has worked in the USA, Japan, Chile, Switzerland, Spain, and South Korea, and holds a Ph.D. from Columbia University.

Programmable Data: Plugging AI Into Your Business Flows

We are generating world record volumes of data every day. We have always generated data, but now we can capture and transmit much more of it. However our machines still do very little to help us understand this data.As things become connected, and human-computer interfaces are changed, we need our machines to understand our data and act on them appropriately.We introduce the novel concept of programmable data and show how Heuro Labs Cognitio transforms unstructured data to primitives that anyone can program to enact intelligent behavior. We will demo applications in industrial IoT, Automotive and Healthcare.

Heuro Labs is a Berlin, based startup that developed Cognitio, a cognitive intelligence framework offered as a service or on premise with special focus on ease of use, privacy, knowledge ownership. Supporting different modalities and data sources, the service enables enterprises to maximize the data value extraction while focusing on high level semantics.

Mohamed Sayed is a technologist with over 20 years of experience working at different layers of the software stack.Most recently he played an instrumental role in building large scale distributed systems at Nokia serving up to a billion users/day handling data of different modalities and velocities. Prior to that, he was at Yahoo! where he was responsible for high trafficked services delivered through thousands of servers in more than 10 global locations. He is the recipient of Nokia Impact Award '11 and Symantec Start award '04

Anomaly Detection in Radiological Images Using Deep Learning

Deep learning is usually linked to Big Data. However, in a medical context radiologists often need help in diagnosing rare conditions with limited historical data. Can Deep Learning prove a useful tool in such cases? COSMONiO is designing NOUS, a deep neural network platform that aims to make high-accuracy predictions using significantly smaller training datasets generated from X-ray, CT, MRI and PET scanners. We will discuss the main challenges and how we address them.

Ioannis is an engineer with a PhD degree in real-time computer vision from Cranfield University, UK. He founded COSMONiO HEALTHLAB in 2015 with a vision to develop intelligent medical technologies based on deep learning. COSMONiO's focus is on designing NOUS, an embedded deep neural network platform that aims to make high-accuracy predictions using significantly smaller training datasets. To achieve this COSMONiO collaborates closely with UMC Groningen, one of the leading university hospitals in The Netherlands. Their research is currently focussing on performing automated anomaly detection on thoracic X-rays.

Alex is a Digital Health Manager at Nuffield Health. With an academic background in Software Engineering at University College London, he is now working on embedding machine learning into health prevention and wellbeing services. Within Nuffield, he advocates personalisation and tailoring across the customer journey. To improve health outcomes, Alex is investigating how behaviour change techniques can be amplified though artificial intelligence.

Generative Biology for Target Discovery: Hypothesis, Novel Target, Novel Small Molecule, In Vitro and In Vivo Validation in Under 18 Months

In November 2016 Insilico Medicine published the first peer-reviewed research paper on the applications of Generative Adversarial Networks (GANs) to the generation of novel molecules with specified properties titled "The Cornucopia of Meaningful Leads: Applying deep adversarial autoencoders for new molecule development in oncology" starting a new field of generative chemistry. Approximately at the same time, it developed the first generative adversarial models for generation of synthetic biological data with the defined properties using age as a generation condition. This approach referred to as generative biology was used for identification of novel molecular targets in a variety of diseases including several liver diseases. In 2018 the first NASH targets were nominated for validation and generative chemistry approach was used to generate novel molecules for these biological targets. In 2019 these targets were validated in vitro and in vivo models. The talk will present the timeline of these discoveries and experimental validation of generative biology and generative chemistry approaches in NASH fibrosis models.

Alex Zhavoronkov, PhD, is the founder and CEO of Insilico Medicine (insilico.com), a leader in next-generation artificial intelligence technologies for drug discovery, biomarker development, and aging research. Since 2015 he invented critical technologies in the field of generative adversarial networks (GANs) and reinforcement learning (RL) for generation of the novel molecular structures with the desired properties and generation of synthetic biological and patient data. He also pioneered the applications of deep learning technologies for prediction of human biological age using multiple data types, transfer learning from aging into disease, target identification, and signaling pathway modeling. Under his leadership Insilico raised over $50 million in 3 rounds and partnered with multiple pharmaceutical, biotechnology, and academic institutions. Prior to founding Insilico, he worked in senior roles at ATI Technologies (acquired by AMD in 2006), NeuroG Neuroinformatics, Biogerontology Research Foundation. Since 2012 he published over 130 peer-reviewed research papers and 2 books including “The Ageless Generation: How Biomedical Advances Will Transform the Global Economy” (Palgrave Macmillan, 2013). He serves on the editorial boards of Aging, Aging Research Reviews, Trends in Molecular Medicine, Frontiers in Genetics, and chairs the Annual Aging Research, Drug Discovery and AI Forum (7th annual in 2020) at Basel Life, one of Europe's largest industry events in drug discovery. Dr. Zhavoronkov holds two bachelor degrees from Queen’s University, a master’s from Johns Hopkins University, and a PhD in Physics and Mathematics from Moscow State University. He is the adjunct professor of artificial intelligence at the Buck Institute for Research on Aging.

Automating Medical Imaging and Preliminary Diagnosis using Virtual Assistants built on multi-functional deep learning networks.

Deep learning has been producing amazing applications in the areas of text, language, speech, images, videos & analytics. However, till now, the applications of Deep Learning always have been singular and point focused. Most of the times in medicine, the interpretations need multi-tasking. For example, while arriving at a diagnosis, a doctor has to verify and gather data sources from various formats like imaging, sensor inputs & feedback from the patient. Point based AI system limit deeper understanding of medical situations. A robot with unified multi-functional deep neural networks can cross communicate the learning and can analyze the situation better. But what are the complexities in developing such networks? What kind of frameworks might work initially? Is it possible to have a modular approach in constructing a multi functional medical assistant?

Arya.ai is a developer platform for artificial intelligence with deep learning tools in language, vision, text, speech, dialogues and reasoning. Using the modules in the platform, developers can build multi-tasking virtual systems that can be customized to a vast number of tasks. Arya.ai has been named recently in the Forbes Asia 30 under 30 list, selected for 'International Innovation awards' by Paris&co - innovation agency from Paris, & has been named as a top 4 next generation technology startup by Silicon Valley Forum. Arya.ai has been working in Deep Learning to automate the AI building process for faster adoption of this technology.

Vinay Kumar is a researcher in Nano-tech from IIT Bombay, developed prediction models in formation of nano-lenses with nano centimeter scale and that has accurately predicted with over 80% accuracy. He founded an AI platform arya.ai along with his co-founder Deekshith Marla who is a researcher in NLP from the same institute. He also authored two books when he was 20y and published them in national and international medium. He also received an excellence award for his research at 21y from ISHRAE.

Dr Sobia Hamid recently wrote a paper for The Wilberforce Society advising policy makers on the opportunities and risks of artificial intelligence in medicine and healthcare. Sobia has been working in the area of personalised medicine and machine learning undertaking scientific and commercial due diligence and marketing for venture capital, biotech and pharma companies. Sobia completed her PhD in Epigenetics at the University of Cambridge, undertaking her research into genomic imprinting at The Babraham Institute. In 2011 Sobia founded Data Insights Cambridge, a networking community for data scientists.

Ferdinando Rodriguez y Baena graduated with a First Class Honours degree in Mechatronics and Manufacturing Systems Engineering from King’s College London in 2000. He then joined the Department of Mechanical Engineering at Imperial College, where he gained a PhD in Medical Robotics. He is now a reader in Mechanical Engineering, where he leads the Mechatronics in Medicine Laboratory.

His past research has focused on development of intra-operative registration methods suitable for minimally invasive knee replacement surgery with a hands-on robot. His work, which was carried out in collaboration with an Imperial College spinout, the Acrobot Company ltd. (which was acquired by Stanmore Implants Worldwide in 2010, and the rights for which were sold to Mako Sugical in 2013), has been presented in over 50 journal publications and peer reviewed conferences, and contributed to the world-first robotic assisted intervention in unicompartmental knee arthroplasty. Dr Rodriguez y Baena’s current research interests lie in the application of mechatronic systems to medicine, in the specific areas of clinical training, diagnostics and surgical intervention.

Guy is a doctor and entrepreneur with an MBA from London Business School who joined the team in January 2015 as Innovation Delivery Lead. He is passionate about the NHS and is driven to maintain its stature as a beacon in the global health economy. His role is to help our partners by catalysing the cultural and operational changes needed to stay competitive in an increasingly modern, digital and patient-centred world, to create opportunities for collaboration and to embed best practice into care pathways.

Over the 15 years prior to joining Imperial College Health Partners he was a Senior Innovation and Growth consultant working around the World within large organisations to launch new ventures while cultivating openness to innovation within the workforce. In parallel he has run his own ventures, which include winning the £1.6m NHS contract to deliver shared-decision making tools so that patients are more involved in decisions about their care, working with the National Clinical Directors and over 300 thought leaders in the NHS.

Detecting Dementia Before it Happens

99.6% of clinical trials for Dementia have not worked. Drugs used today do not address the root causes but only alleviate symptoms for a few years. To discover drugs that work, pharmaceutical companies have shifted clinical trial towards earlier stage patients. However, the putative early stage of Dementia, Mild Cognitive Impairment (MCI), progresses to actual Dementia at less than 15% per year, meaning a clinical trial needs to enrol many patients and last many years to show effectiveness. We have developed algorithms that can find the patients who convert next year with an accuracy of 75%, meaning that clinical trials need fewer patients for less time, drastically decreasing cost.

Alejandro (Sasha) Vicente Grabovetsky is a Cognitive Neuroscientist specialising in memory. He graduated with a BA and a PhD from Cambridge University and did his Post-Doc at the Donders Centre for Cognitive Neuroimaging. He left academia and joined the Entrepreneur First program in London. There, he met his co-founder, livier van den Biggelaar, a Software Engineer who built complex Machine Learning systems in his PhD at the University of Brussels and in industry.

Predictive Algorithms and their Impact in Chronic Diseases Management: Diabetes Case

Eduardo will share his vision on the major role that artificial intelligence can have in the digital transformation of the healthcare sector. As a medical doctor and CEO of a medical devices company, he has negotiated with major stakehoders and has clear insights about the needs of both the industry and the patient. Personalized self care, preventive diagnosis, individual treatments and patient support programs, everything available thanks to predictive models and easy interfaces.

Eduardo W. Jorgensen: 2015 UAM Bachelor in Medicine, Built several teams for medical research doing college, passionate about medicine and new technologies, specially in the area of neuroscience and perception. My goal is to change the wy we live with chronic diseases. Spanish national with great interest in medicine, neuroscience and technologies. Perfect domain of English for fluent conversations and negotiations. Had an enrollment of honor prior to college, got into medical school in 2009 and graduated in 2015. In the period, got selected for StartupsMansion (NYC entrepreneurs program), had funded two companies: Livinplans (Surprise travelling agency) and MedicSen (non invasive artificial Pancreas for diabetes) and have been accelerated by TURN8 in Dubai. At the end of the day, I try to learn as much as possible about the world and get surrounded by the top innovative people to form good multidisciplinary teams capable of disrupting every aspect.

Optimising CRISPR Genome Editing using Machine Learning

Desktop Genetics is designing and testing CRISPR experiments to create training datasets for an AI that can be used for genome engineering applications in microbe, plant, animal and human cells. Decision rules are commonly used to design CRISPR genome editing procedures. Such rules use sequence-level and contextual features of the predicted genomic cut site locations to predict the activity, specificity, and outcome of using CRISPR in vitro and in vivo. We demonstrate significant improvements in designing genome editing procedures, which we have tested in an immortalized melanoma cell line and report on here.

As a biochemist turned software engineer, Riley recognized the enormous untapped potential for software innovation in the life sciences. An alum of Genentech, he brings ten years of laboratory-based genetic engineering experience to Desktop Genetics, where his team's research is focused on software systems for the rational design of CRISPR/Cas9 Genome Editing vectors and libraries. Riley is passionate about developing the next-generation of bioinformatics tools that drive science forward. He holds a Bachelor's of Engineering from Dartmouth College, a Bachelor’s of Arts in Biochemistry from Colby College, and and an MPhil in Bioscience Enterprise from the University of Cambridge.

Deep learning for healthcare technologies

With the confluence of very large datasets now becoming available in healthcare along with methods from deep learning to providing new ways of using them, healthcare is poised to adopt new tools based on such technologies in coming years. This talk describes work undertaken between the University of Oxford and the Oxford University Hospitals in developing early systems in this field.

David A. Clifton is Associate Professor of Engineering Science at the University of Oxford and leads the Computational Health Informatics (CHI) Laboratory, after having trained in information engineering at the University of Oxford. CHI Lab focusses on the interface between machine learning and healthcare, in partnership with leading clinicians from the Oxford University Hospitals NHS Trust, and has grown rapidly to over 20 members in the two years since its founding, with support from the Wellcome Trust, UK Department of Health, NHS National Institute of Health Research, Engineering & Physical Sciences Research Council, Royal Academy of Engineering, Natural Environment Research Council, and the Bill & Melinda Gates Foundation.